We are developing a superior model for stroke to investigate this process and evaluate therapies. Awake monkeys are studied to avoid the confusing effects of anesthesia. The middle cerebral artery (MCA) is blocked temporarily with an implanted ligature to cause local brain ischemia. We then study neurologic function, cerebral blood flow, and eventual brain damage. Results to date show that: (1) variable collateral blood supply is responsible for the variable brain damage in experimental strokes. (2) There is a threshold for infarction, below which irreversible damage occurs in two to four hours. This important new finding will allow tests of stroke treatment. (3) The unanesthetized monkey model is excellent for testing treatment, but the inherent variability in alertness means the model cannot detect small changes in brain blood flow--ancillary studies under anesthesia will be needed. These encouraging findings set the stage for studies of stroke treatment.